After COVID-19, some patients develop long-term symptoms. Whether such symptoms correlate with immune responses, and how long immunity persists, is not yet clear. This study focused on mild COVID-19 and investigated correlations of immunity with persistent symptoms and immune longevity. Persistent complications, including headache, concentration difficulties and loss of smell/taste, were reported by 51 of 83 (61%) participants and decreased over time to 28% one year after COVID-19. Specific IgA and IgG antibodies were detectable in 78% and 66% of participants, respectively, at a 12-month follow-up. Median antibody levels decreased by approximately 50% within the first 6 months but remained stable up to 12 months. Neutralizing antibodies could be found in 50% of participants; specific INFgamma-producing T-cells were present in two thirds one year after COVID-19. Activation-induced marker assays identified specific T-helper cells and central memory T-cells in 80% of participants at a 12-month follow-up. In correlative analyses, older age and a longer duration of the acute phase of COVID-19 were associated with higher humoral and T-cell responses. A weak correlation between long-term loss of taste/smell and low IgA levels was found at early time points. These data indicate a long-lasting immunological memory against SARS-CoV-2 after mild COVID-19.

Memory T-cell responses following infection with coronaviruses are reportedly long-lived and provide long-term protection against severe disease. Whether vaccination induces similar long-lived responses is not yet clear since, to date, there are limited data comparing memory CD4+ T-cell responses induced after SARS-CoV-2 infection versus following vaccination with BioNTech/Pfizer BNT162b2. We compared T-cell immune responses over time after infection or vaccination using ELISpot, and memory CD4+ T-cell responses three months after infection/vaccination using activation-induced marker flow cytometric assays. Levels of cytokine-producing T-cells were remarkably stable between three and twelve months after infection, and were comparable to IFNγ+ and IFNγ+IL-2+ T-cell responses but lower than IL-2+ T-cell responses at three months after vaccination. Consistent with this finding, vaccination and infection elicited comparable levels of SARS-CoV-2 specific CD4+ T-cells after three months in addition to comparable proportions of specific central memory CD4+ T-cells. By contrast, the proportions of specific effector memory CD4+ T-cells were significantly lower, whereas specific effector CD4+ T-cells were higher after infection than after vaccination. Our results suggest that T-cell responses—as measured by cytokine expression—and the frequencies of SARS-CoV-2-specific central memory CD4+T-cells—indicative of the formation of the long-lived memory T-cell compartment—are comparably induced after infection and vaccination.